Although epilepsy has a large genetic contribution, causative genes are not known for most epilepsy patients. And while epilepsy can be inherited in families, it is more commonly 'sporadic' or spontaneous, without a clear family history, and causes of sporadic epilepsy cases are mostly unknown. Since the cause of epilepsy often directs the choice of treatments, categorizing patients is of great importance therapeutically. Many epilepsy syndromes appear to involve 'somatic' mutations, in which a subpopulation of brain cells shows a mutation not shared by all cells of the body, because the mutation occurred during mitosis of somatic cells of the embryo after fertilization. Our lab and other have reported on somatic mutations is several genes known to cause epilepsy, however until now, technological limitations have prevented a systematic search for somatic mutations in epilepsy. Whole genome or whole exome sequencing will not identify somatic mutations unless the study is designed to detect them: responsible mutations may not be present in most blood cells, but would instead be limited to cells in the brain. Even relevant mutations limited to neurons may be missed by bulk brain sequencing because 1] neurons are surrounded by glial cells with distinct embryological origins, and with whom they would not be expected to share mutations, and 2] neurons themselves are derived from two embryological origins, with pyramidal neurons derived from cortical proliferative regions, and inhibitory nonpyramidal neurons migrating into cortex from distant subcortical regions, so that bulk sequencing of postmortem brain tissue may also not detect mutations limited to specific neuronal populations. We have recently developed technology that allows sorting of single or small numbers of cerebral cortical neuronal nuclei, and amplification of their genomes in quantities sufficient for any next-generation sequencing. We are proposing to use this technology to examine cortical neurons from postmortem brains or cortical resections from epilepsy patients in order to 1) perform a systematic assessment of copy number variation (CNV) in cerebral neurons; and 2) perform whole exome sequencing to identify mutations in known epilepsy genes. Comparison of single neuron sequence to that from other cells of the body could then determine how frequently spontaneous mutations occur in cortical neurons, and identify and catalogue these mutations. Our proposed study thus enables a systematic analysis of all mechanisms of somatic mutation in the epileptic human brain.